Two Dimensionalism about Other Minds, and Its Implications for Brain Organoids and Robots

You know (I hope!) that you are conscious. How do you know that other people are conscious too? This is the classic "problem of other minds".

The question isn't mainly developmental or psychological, but epistemic: What justifies you in believing that others have conscious experiences like yours -- feelings of joy and pain, thoughts in inner speech, dreams, sensory experiences -- instead of being, so to speak, automata who are all dark inside?

One common answer appeals to analogy: You are justified on grounds of others' similarity to you. It would be strange if entities so behaviorally and physiologically similar didn't also have similar streams of inner experience.

John Stuart Mill expresses it thus:

By what evidence do I know, or by what considerations am I led to believe, that there exist other sentient creatures; that the walking and speaking figures which I see and hear, have sensations and thoughts, or in other words, possess Minds?... I conclude that other human beings have feelings like me, because, first, they have bodies like me, which I know, in my own case, to be the antecedent condition of feelings; and because, secondly, they exhibit the acts, and other outward signs, which in my own case I know by experience to be caused by feelings (An Examination of Sir William Hamilton's Philosophy, 3rd ed., 1867, p. 237).

Notice that Mill appeals to two very different types of similarity: similarity of body and similarity of acts and outward signs.

[title page of John Stuart Mill, An Examination of Sir William Hamilton's Philosophy, 3rd edition]

In a recent paper, Ned Block makes a similar distinction between first order realizer properties, like being made of a certain kind of "meat", and second order functional role properties, like being the kind of thing that causes crying.

Block's functionalist jargon would have been unfamiliar to Mill, but the idea is much the same. Mill writes:

I am conscious in myself of a series of facts connected by a uniform sequence, of which the beginning is modifications of my body, the middle is feelings, the end is outward demeanor. In the case of other human beings I have the evidence of my senses for the first and last links of the series, but not for the intermediate link; which must either the be same in others as in myself, or a different one.... (p. 237-238).

Mill, like a good functionalist, seeks something to fill the middle link of a causal chain from cause to X to effect. The filler or "realizer" of this functional role property could potentially be anything, though in him it is a feeling.

For example, in me, a mosquito sting and the resulting red bump leads to a feeling of itchiness, which in turn leads to scratching. In others, I see the same sting and bump and the same scratching, but I cannot see the itchiness between.

At first, Mill suggests that it's reasonable to assume the intermediate feeling (the itchiness) simply on the grounds that "no other force need be supposed" (p. 238). But later he supports the claim also by appeal to physiological similarity:

I look about me, and though here is only one... body... which is connected with all my sensations in this peculiar manner, I observe that there is a great multitude of other bodies, closely resembling in their sensible properties... this particular one, but whose modifications do not call up, as those of my own body do, a world of sensations in my consciousness. Since they do not do so in my consciousness, I infer that they do it out of my consciousness, and that to each of them belongs a world of consciousness of its own... (p. 238-239).

Because others' bodies are like mine, I infer that the intermediate X -- the feeling of itchiness, in our example -- is also similar.

Let's call this view two-dimensionalism about other minds: Only when another entity is both physiologically and functionally (that is, in terms of typical causes and effects) similar to me am I justified in inferring that it has experiences like mine. When the two dimensions diverge, skepticism follows.

Human babies are physiologically similar to adult humans but functionally quite different. In the bad old days, I gather, there used to be doubts about whether babies were conscious, for example, whether they could actually feel pain (and thus anesthesia was not regularly practiced). Yet because the causes and effects of their pain responses are similar, as well as their physiology, such doubt was misplaced.

Brain organoids are a more difficult case. Human brain cells can be grown in vitro, in clusters of tens of millions of neurons. Could consciousness arise in such systems? Functionally, brain organoids are radically impoverished compared to ordinary humans. But if what matters is neurophysiology, maybe a sufficiently large or well-structured brain organoid would be conscious.

Robots present a complementary case: Language models are becoming similar to us in linguistic behavior. We might guess or imagine that some future robots will become functionally or behaviorally similar to us in other ways too, while remaining physiologically very different. Block argues in his recent paper, as well as in earlier work, that we don't know that the physiology doesn't matter. Maybe only "meat machines" can be conscious, while silicon machines, even if functionally very similar to us, could never be conscious.

The crux of the matter lies, perhaps, in whether two-dimensionalism or one-dimensionalism is the right response to the problem of other minds. The one-dimensionalist -- Mill, briefly -- holds that if we see the right types of similar causal relationships between inputs and outputs, that's enough to justify attributing consciousness (perhaps on grounds of simplicity or parsimony: "no other force need be supposed"). The two-dimensionalist, like Block, thinks doubt is justified unless there's both functional and physiological similarity.

Two dimensionalists are thereby committed to doubting AI consciousness, unless we someday create AI that is not only functionally but physiologically similar to us.

Must one-dimensionalist functionalists reject organoid consciousness? That's not as clear. I see at least two paths for them to accept organoid consciousness. First, they might define the functional roles in terms of features internal to neural systems -- not mosquito bites and scratching, but things like information sharing across a global workspace. Second, they might use the functional role to identify a physiological type, and then a la David Lewis, attribute consciousness whenever that physiological type is present, even if it isn't -- in that particular system -- playing its typical functional role.

New Book in Draft: AI and Consciousness

This book is a skeptical overview of the literature on AI consciousness.

We will soon create AI systems that are conscious according to some influential, mainstream theories of consciousness but are not conscious according to other influential, mainstream theories of consciousness. We will not be in a position to know which theories are correct and whether we are surrounded by AI systems as richly and meaningfully conscious as human beings or instead only by systems as experientially blank as toasters. None of the standard arguments either for or against AI consciousness take us far.

Table of Contents

Chapter One: Hills and Fog
Chapter Two: What Is Consciousness? What Is AI?
Chapter Three: Ten Possibly Essential Features of Consciousness
Chapter Four: Against Introspective and Conceptual Arguments for Essential Features
Chapter Five: Materialism and Functionalism
Chapter Six: The Turing Test and the Chinese Room
Chapter Seven: The Mimicry Argument Against AI Consciousness
Chapter Eight: Global Workspace Theories and Higher Order Theories
Chapter Nine: Integrated Information, Local Recurrence, Associative Learning, and Iterative Natural Kinds
Chapter Ten: Does Biological Substrate Matter?
Chapter Eleven: The Problem of Strange Intelligence
Chapter Twelve: The Leapfrog Hypothesis and the Social Semi-Solution

Draft available here.

Per my usual custom, anyone who gives comments on the entire manuscript (by email please, to my academic address at ucr.edu) will receive not only the usual acknowledgement but an appreciatively signed copy once it appears in print.

Are Weird Aliens Conscious? Three Arguments (Two of Which Fail)

Most scientists and philosophers of mind accept some version of what I'll call "substrate flexibility" (alternatively "substrate independence" or "multiple realizability") about mental states, including consciousness. Consciousness is substrate flexible if it can be instantiated in different types of physical system -- for example in a squishy neurons like ours, in the silicon chips of a futuristic robot, or in some weird alien architecture, carbon based or not.

Imagine we encounter a radically different alien species -- one with a silicon-based biology, perhaps. From the outside, they seem as behaviorally sophisticated as we are. They build cities, fly spaceships, congregate for performances, send messages to us in English. Intuitively, most of us would be inclined to say that yes, such aliens are conscious. They have experiences. There is "something it's like" to be them.

But can we argue for this intuition? What if carbon is special? What if silicon just doesn't have the je ne sais quoi for consciousness?

This kind of doubt isn't far fetched. Some people are skeptical of the possibility of robot consciousness on roughly these grounds, and some responses to the classic "problem of other minds" rely on our biological as well as behavioral similarity to other humans.

If we had a well-justified universal theory of consciousness -- one that applies equally to aliens and humans -- we could simply apply it. But as I've argued elsewhere, we don't have such a theory and we likely won't anytime soon.

Toward the conclusion that behaviorally sophisticated aliens would be conscious regardless of substrate, I see three main arguments, two of which fail.

Argument 1: Behavioral Sophistication Is Best Explained by Consciousness

The thought is simple. These aliens are, by hypothesis, behaviorally sophisticated. And the best explanation for sophisticated behavior is that they have inner conscious lives.

There are two main problems with this argument.

First, unconscious sophistication. In humans, unconscious behavior often displays complexity without consciousness. Bipedal walking requires delicate, continuous balancing, quickly coordinating a variety of inputs, movements, risks, and aims -- mostly nonconscious. Expert chess players make rapid judgments they can't articulate, and computers beat those same experts without any consciousness at all.

Second, question-begging. This argument simply assumes what the skeptic denies: that the best explanation for alien behavior is consciousness. But unless we have a well justified, universally applicable account of the difference between conscious and unconscious processing -- which we don't -- the skeptic should remain unmoved.

Argument 2: The Functional Equivalent of a Human Could Be Made from a Different Substrate

This argument has two steps:

(1.) A functional equivalent of you could be made from a different substrate.

(2.) Such a functional equivalent would be conscious.

One version is David Chalmers' gradual replacement or "fading qualia" argument. Imagine swapping your neurons, one by one, with silicon chips that are perfect functional equivalents. If this process is possible, Premise 1 is true.

In defense of Premise 2, Chalmers appeals to introspection: During the replacement, you would notice no change. After all, if you did notice a change, that would presumably have downstream effects on your psychology and/or behavior, so functional equivalence would be lost. But if consciousness were fading away, you should notice it. Since you wouldn't, the silicon duplicate must be conscious.

Both premises face trouble.

Contra Premise 1, as Rosa Cao, Ned Block, Peter Godfrey-Smith and others have argued, it is probably not possible to make a strict functional duplicate out of silicon. Neural processing is subserved by a wide variety of low level mechanisms -- for example nitric oxide diffusion -- that probably can't be replicated without replicating the low-level chemistry itself.

Contra Premise 1, as Ned Block and I have argued, there's little reason to trust introspection in this scenario. If consciousness did fade during the swap, whatever inputs our introspective processes normally rely on will be perfectly mimicked by the silicon replacements, leaving you none the wiser. This is exactly the sort of case where introspection should fail.

[DON'T PANIC! It's just a weird alien (image source)]

Argument 3: The Copernican Argument for Alien Consciousness

This is the argument I favor, developed in a series of blog posts and a paper with Jeremy Pober. According to what Jeremy and I call The Copernican Principle of Consciousness, among behaviorally sophisticated entities, we are not specially privileged with respect to consciousness.

This basic thought is, we hope, plausible on its face. Imagine a universe with at least a thousand different behaviorally sophisticated species, widely distributed in time and space. Like us, they engage in complex, nested, long-term planning. Like us, they communicate using sophisticated grammatical language with massive expressive power. Like us, they cooperate in complex, multi-year social projects, requiring the intricate coordination of many individuals. While in principle it's conceivable that only we are conscious and all these other species are merely nonconscious zombies, that would make us suspiciously special, in much the same way it would be suspiciously special if we happened to occupy the exact center of the universe.

Copernican arguments rely on a principle of mediocrity. Absent evidence to the contrary, we should assume we don't occupy a special position. If we alone were conscious, or nearly alone, we would occupy a special position. We'd be at the center of the consciousness-is-here map, so to speak. But there's no reason to think we are lucky in that way.

Imagine a third-party species with a consciousness detector, sampling behaviorally sophisticated species. If they find that most or all such species are conscious, they won't be surprised when they find that humans, too, are conscious. But if species after species failed, and then suddenly humans passed, they would have to say, "Whoa, something extraordinary is going on with these humans!" It's that kind of extraordinariness that Copernican mediocrity tells us not to expect.

Why do we generally think that behaviorally sophisticated weird aliens would be conscious? I don't think the core intuition is that you need consciousness to explain sophistication or that the aliens could be functionally exactly like us. Rather, the core intuition is that there's no reason to think neurons are special compared to any other substrate that can support sophisticated patterns of behavior.

Three Epistemic Problems for Any Universal Theory of Consciousness

By a universal theory of consciousness, I mean a theory that would apply not just to humans but to all non-human animals, all possible AI systems, and all possible forms of alien life. It would be lovely to have such a theory! But we're not at all close.

This is true sociologically: In a recent review article, Anil Seth and Tim Bayne list 22 major contenders for theories of consciousness.

It is also true epistemically. Three broad epistemic problems ensure that a wide range of alternatives will remain live for the foreseeable future.

First problem: Reliance on Introspection

We know that we are conscious through, presumably, some introspective process -- through turning our attention inward, so to speak, and noticing our experiences of pain, emotion, inner speech, visual imagery, auditory sensation, and so on. (What is introspection? See my SEP encyclopedia entry Introspection and my own pluralist account.)

Our reliance on introspection presents three methodological challenges for grounding a universal theory of consciousness:

(A.) Although introspection can reliably reveal whether we are currently experiencing an intense headache or a bright red shape near the center of our visual field, it's much less reliable about whether there's a constant welter of unattended experience or whether every experience comes with a subtle sense of oneself as an experiencing subject. The correct theory of consciousness depends in part on the answer to such introspectively tricky questions. Arguably, these questions need to be settled introspectively first, then a theory of consciousness constructed accordingly.

(B.) To the extent we do rely on introspection to ground theories of consciousness, we risk illegitimately presupposing the falsity of theories that hold that some conscious experiences are not introspectable. Global Workspace and Higher-Order theories of consciousness tend to suggest that conscious experiences will normally be available for introspective reporting. But that's less clear on, for example, Local Recurrence theories, and Integrated Information Theory suggests that much experience arises from simple, non-introspectable, informational integration.

(C.) The population of introspectors might be much narrower than the population of entities who are conscious, and the first group might be unrepresentative of the latter. Suppose that ordinary adult human introspectors eventually achieve consensus about the features and elicitors of conscious in them. While indeed some theories could thereby be rejected for failing to account for ordinary human adult consciousness, we're not thereby justified in universalizing any surviving theory -- not at least without substantial further argument. That experience plays out a certain way for us doesn't imply that that it plays out similarly for all conscious entities.

Might one attempt a theory of consciousness not grounded in introspection? Well, one could pretend. But in practice, introspective judgments always guide our thinking. Otherwise, why not claim that we never have visual experiences or that we constantly experience our blood pressure? To paraphrase William James: In theorizing about human consciousness, we rely on introspection first, last, and always. This centers the typical adult human and renders our grounds dubious where introspection is dubious.

Second problem: Causal Confounds

We humans are built in a particular way. We can't dismantle ourselves and systematically tweak one variable at a time to see what causes what. Instead, related things tend to hang together. Consider Global Workspace and Higher Order theories again: Processes in the Global Workspace might almost always be targeted by higher order representations and vice versa. The theories might then be difficult to empirically distinguish, especially if each theory has the tools and flexibility to explain away putative counterexamples.

If consciousness arises at a specific stage of processing, it might be difficult to rigorously separate that particular stage from its immediate precursors and consequences. If it instead emerges from a confluence of processes smeared across the brain and body over time, then causally separating essential from incidental features becomes even more difficult.

Third problem: The Narrow Evidence Base

Suppose -- very optimistically! -- that we figure out the mechanisms of consciousness in humans. Extrapolating to non-human cases will still present an intimidating array of epistemic difficulties.

For example, suppose we learn that in us, consciousness occurs when representations are available in the Global Workspace, as subserved by such-and-such neural processes. That still leaves open how, or whether, this generalizes to non-human cases. Humans have workspaces of a certain size, with a certain functionality. Might that be essential? Or would literally any shared workspace suffice, including the most minimal shared workspace we can construct in an ordinary computer? Human workspaces are embodied in a living animal with a metabolism, animal drives, and an evolutionary history. If these features are necessary for consciousness, then conclusions about biological consciousness would not carry over to AI systems.

In general, if we discover that in humans Feature X is necessary and sufficient for consciousness, humans will also have Features A, B, C, and D and lack Features E, F, G, and H. Thus, what we will really have discovered is that in entities with A, B, C, and D and not E, F, G, or H, Feature X is necessary and sufficient for consciousness. But what about entities without Feature B? Or entities with Feature E? In them, might X alone be insufficient? Or might X-prime be necessary instead?

The obstacles are formidable. If they can be overcome, that will be a very long-term project. I predict that new theories of consciousness will be added faster than old theories can be rejected, and we will discover over time that we were even further away from resolving these questions in 2025 than we thought we were.

[a portion of a table listing theories of consciousness, from Seth and Bayne 2022]

Types and Degrees of Turing Indistinguishability; Thinking and Consciousness

Types and Degrees of Indistinguishability

The Turing test (introduced by Alan Turing in a 1950 article) treats linguistic indistinguishability from a human as sufficient grounds to attribute thought (alternatively, consciousness) to a machine. Indistinguishability, of course, comes in degrees.

In the original setup, a human and a machine, through text-only interface, each try to convince a human judge that they are human. The machine passes if the judge cannot tell which is which. More broadly, we might say that a machine "passes the Turing test" if its textual responses strike users as sufficiently humanlike to make the distinction difficult.

[Alan Turing in 1952; image source]

Turing tests can be set with a relatively low or high bar. Consider a low-bar test:

* The judges are ordinary users, with no special expertise.
* The interaction is relatively brief -- maybe five minutes.
* The standard of indistinguishability is relaxed -- maybe if 20% of users guess wrong, that suffices.

Contrast that with a high-bar test:

* The judges are experts in distinguishing humans from machines.
* The interaction is relatively long -- an hour or more.
* The standard of indistinguishability is stringent -- if even 55% of judges guess correctly, the machine fails.

The best current language models already pass a low-bar test. But it will be a long time before language models pass this high-bar test, if they ever do. So let's not talk about whether machines do or not pass "the" Turing test. There is no one Turing test.

The better question is: What type and degree of Turing-indistinguishability does a machine possess? Indistinguishability to experts or non-experts? Over five minutes or five hours? With what level of reliability?

We might also consider topic-based or tool-relative Turing indistinguishability. A machine might be Turing indistinguishable (to some judges, for some duration, to some standard) when discussing sports and fashion, but not when discussing consciousness, or vice versa. It might fool unaided judges but fail when judges employ AI detection tools.

Turing himself seems to have envisioned a relatively low bar:

I believe that in about fifty years' time it will be possible, to programme computers... to make them play the imitation game so well that an average interrogator will not have more than 70 per cent chance of making the right identification after five minutes of questioning (Turing 1950, p. 442)

I've bolded Turing's implied standards of judge expertise, indistinguishability threshold, and duration.

What bar should we adopt? That depends on why we care about Turing indistinguishability. For a customer service bot, indistinguishability by ordinary people across a limited topic range for brief interaction might suffice. For an "AI girlfriend", hours of interaction might be expected, with occasional lapses tolerated or even welcomed.

Turing Tests for Real Thinking and Consciousness?

But maybe you're interested in the metaphysics, as I am. Does the machine really think? Is it really conscious? What kind and degree of Turing indistinguishability would establish that?

For thinking, I propose that when it becomes practically unavoidable to treat the machine as if it has a particular set of beliefs and desires that are stable over time, responsive to its environment, and idiosyncratic to its individual state, then we might as well say that it does have beliefs and desires, and that it thinks. (My own theory of belief requires consciousness for full and true belief, but in such a case I don't think it will be practical to insist on this.)

Current language models aren't quite there. Their attitudes lack sufficient stability and idiosyncrasy. But a language model integrated into a functional robot that tracks its environment and has specific goals would be a thinker in this sense. For example: Nursing Bot A thinks the pills are in Drawer 1, but Nursing Bot B, who saw them moved, knows that they're in Drawer 2. Nursing Bot A would rather take the long, safe route than the short, riskier route. We will want attribute sometimes true, sometimes false environment-tracking beliefs and different stable goal weightings. Belief, desire, and thought attribution will be too useful to avoid.

For consciousness, however, I think we should abandon a Turing test standard.

Note first that it's not realistic to expect any machine ever to pass the very highest bar Turing test. No machine will reliably fool experts who specialize in catching them out, armed with unlimited time and tools, needing to exceed 50% accuracy by only the slimmest margin. To insist on such a high standard is to guarantee that no machine could ever prove itself conscious, contrary to the original spirit of the Turing test.

On the other hand, given enough training and computational power, machines have proven to be amazing mimics of the superficial features of human textual outputs, even without the type of underlying architecture likely to support a meaningful degree of consciousness. So too low a bar is equally unhelpful.

Is there reason to think that we could choose just the right mid-level bar -- high enough to rule out superficial mimicry, low enough not to be a ridiculously unfair standard?

I see no reason to think there must be some "right" level of Turing indistinguishability that reliably tests for consciousness. The past five years of language-model achievements suggest that with clever engineering and ample computational power, superficial fakery might bring a nonconscious machine past any reasonable Turing-like standard.

Turing never suggested that his test was a test of consciousness. Nor should we. Turing indistinguishability has potential applications, as described above. But for assessing consciousness, we'll want to look beyond outward linguistic behavior -- for example, to interior architecture and design history.

When Is a Theory Superficial?

by Jeremy Pober and Eric Schwitzgebel

Twelve years ago, one of us (ES) distinguished two kinds of theories: superficial and deep. Nearly any phenomenon can be approached in a superficial or deep manner. A superficial judge of human beauty treats it as skin deep. A superficial reading of Shakespeare takes characters at their word and focuses on the obvious aspects of each scene. A superficial housecleaning ignores the backsides and undersides of household items.

And of course one can have a superficial theory of belief. Phenomenal dispositionalism is intended to be such a theory. According to phenomenal dispositionalism, whether someone believes that P is a matter of whether they have certain behavioral, phenomenal (i.e., experiential), and cognitive dispositions, specifically, the dispositions that are "stereotypical" of a person who believes that P. Compare: To be an extravert just is to have the behavioral, phenomenal, and cognitive dispositions stereotypical of extraversion.

Superficial theories contrast with deep theories. Among theories of belief, the main contrast has been with the computationalist, representationalist functionalism made famous by Jerry Fodor (1987) and recently defended by Jake Quilty-Dunn and Eric Mandelbaum.

But what makes a theory of some property P superficial (or deep)? Twelve years ago, ES offered an answer: It depends on the theory's relationship to surface properties. Surface properties are observable features of a phenomenon that a theory of P is designed to explain (in a loose sense of "observable"[1]).

What relation to surface properties must a theory have to be superficial or deep? Back in 2013, ES said that "relative to a class of surface phenomena... a property is superficial if it identifies possession of the property simply with patterns in the surface phenomena" (2013, 77). And a theory is deep "relative to a class of surface phenomena... if it identifies possession of the property with some feature other than patterns in those same surface phenomena -- some feature that presumably explains or causes or underwrites those surface patterns" (ibid.).

This definition fits our toy examples above. A superficial judge of beauty relies on the most easily observable physical patterns, a superficial reading of Shakespeare focuses on surface-level dialogue, and a superficial house-cleaning treats looking clean as clean.

However, we have reason to be unsatisfied with this definition. [ES thanks JP for emphasizing this point in a series of discussions.]

Consider poison, a "causal concept" in David Armstrong (1968)'s sense: a concept defined by its causes and/or effects. Poison can be defined in terms of biologically harming a person when ingested (with refinements to differentiate poisoning from, say, drinking lava).[2] If I explain a death by saying that a person was poisoned, you can infer that the death was caused by ingestion rather than, say, hypothermia. That's informative -- but much less informative than saying that the person ingested cyanide, because chemical types like cyanide are defined structurally, allowing detailed explanations of how they interact with human physiology.

A theory of health that only has non-structural causal concepts like "poison" (or "medicine") would be a superficial theory of health. A deep theory, in contrast, invokes underlying mechanisms.

Yet, by ES's 2013 definition, a theory appealing to poison wouldn't count as superficial, because ingesting poison isn't merely related to death as two parts of a superficial pattern. Poison causes death.[3]

In a new draft, ES proposes a revised definition: a theory of property P is superficial if "whether an entity has property [P] is determined (that is, constituted or grounded...) entirely by superficial facts about that entity", where superficial facts are readily observed facts. For causal concepts, being the cause of is a constitutive relationship. This new definition thus accommodates causal superficialism, where poisons cause death and medicines cause recoveries, as inferable from readily observable relationships (such as randomized controlled trials), without appeal to deeper structural features.

That's a good thing! Otherwise, phenomenal dispositionalism only counts as a superficial theory of belief if dispositions don't cause their manifestations. Some philosophers of mind (e.g., Ryle 1949) indeed view dispositions non-causally. But others, like Armstrong (1968), propose a "realist" conception: Dispositions are type-identical to their causal bases. Fragility, for example, is identified with the microstructural features that cause fragile objects to break when struck.[4]

In his original articulation of phenomenal dispositionalism, ES expressed willingness to accept such a realist view (2002, 273n18). This version of dispositionalism can be considered equivalent to a version of functionalism (which holds that mental states can be defined in terms of their causal relations to inputs, outputs, and other mental states). Georges Rey (1997) calls this type of functionalism superficial functionalism, where all functional/causal roles are defined only in relation to behavior, thought, experience, and "similar" states (e.g., desire is similar to belief, so a superficial functionalist theory of belief can include relations to desires).[5]

Of course, deep theories also often employ causal explanations. So if causal superficial theories are possible, what distinguishes them from deep theories? The answer is that causal posits in superficial theories have minimal explanatory content, whereas deep theories have excess explanatory content.[6] Posits with minimal explanatory content explain all that they were posited to explain and no more, whereas posits with excess content make further falsifiable predictions.

Consider the difference between a geneticist working right after Gregor Mendel published his work on heritability, and one working after Franklin, Watson, and Crick had mapped the structure of DNA and demonstrated how it instantiated genetic material. Mendel's theory, which gives us the posits of trait, gene, allele, and dominant/recessive, is a powerful theory (much like belief/desire psychology), but it doesn't explain how genes and alleles have the properties that they do. An allele is just the genetic material for a variant in phenotype, e.g., blood type A versus B or O. But in the initial Mendelian framework, it was defined as "whatever is responsible for variance in (e.g.) blood type".

[illustration of Mendel's superficial causal theory; image source]

Contrast with someone working in the latter half of the 20th century. They know that genetic information is realized in DNA (& RNA), which via its repeating base patterns and double helix structure, acts as a base code for the information that constitutes alleles. In other words, they know how genes carry genetic information.[7]

Superficial theories needn't be acausal, but if they posit causal relationships, those relationships must exist among the readily observable features, without invoking hidden structures or mechanisms that yield additional explanatory content. In contrast, the later 20th century theory makes many more falsifiable predictions -- those that follow from the structure of DNA -- and thus has excess explanatory content.

--------------------------------------------

[1] This might not match the sense of "observable" sometimes used in philosophy of science. Dennett (1994) defines observable from his perspective of "urbane verificationism" and, for a theory of attitudes, takes the same list of surface properties to be observable as ES: behavior, thought, and experience.

[2] More precisely, poison is always a two-place predicate, poison-for-S where S is some group of organisms such as a species. When no such group is specified, we can treat instances of poison as poison-for-humans. We are ignoring contact poisons and other complications.

[3] Thus the distinction between superficial and deep theories is not a distinction about noncausal versus causal explanations. Consequently, the superficial/deep distinction as applied to the attitudes does not end up reducing to Devin Curry's distinction between beliefs as properties of persons and beliefs as "cogs" of cognitive science (Curry 2021).

[4] The standard way of defining a causal basis is in terms of physical properties, such as microstructural properties defining "fragility". However this is not a strict requirement. One can posit a mental kind (as in Quilty-Dunn and Mandelbaum 2018 where representations are the causal bases of dispositions constitutive of belief stereotypes) or even a higher-order kind (as in Prior, Pargetter, and Jackson 1982).

[5] Rey (1994; 1997) invokes this term in a debate with Dan Dennett that parallels the debate between ES and Quilty-Dunn and Mandelbaum. While the overall debate turns on different issues, the definition of superficialist theories of belief lines up. Examples of this sort of functionalism plausibly include David Armstrong (1968), the David Lewis of "An Argument for the Identity Theory" (1966) but maybe not the David Lewis of "Mad Pain and Martin Pain" (1980), and Adam Pautz 2021).

[6] Term adopted from Lakatos's (1968) notion of "excess" explanatory content.

[7] The DNA example also lets us talk about different levels or degrees of depth. The late 20th century theory of a gene is a deep one, but so is a theory mid-way between that and Mendel's. In the first years of the 20th century scientists identified chromosomes as the realizer of genes, but did not know that chromosomes were made of DNA (they thought they were proteins). This theory too is deep -- there are excess predictions made by the assignment of genetic material to chromosomes -- but not as deep as later views, because not nearly as many excess predictions were made. We can tentatively call such a theory formally deep, whereas a theory that more fully explains how the posit in question (genes, beliefs) has the properties that it does is substantively deep.

Elisabeth of Bohemia 1, Descartes 0

I'm loving reading the 1643 correspondence between Elisabeth of Bohemia and Descartes! I'm embarrassed to confess that I hadn't read it before now; the standard Cottingham et al. edition presents only selections from Descartes' side. I'd seen quotes of Elisabeth, but not the whole exchange as it played out. Elisabeth's letters are gems. She has Descartes on the ropes, and she puts her concerns so plainly and sensibly (in Bennett's translation; I haven't attempted to read the antique French). You can practically feel Descartes squirming against her objections. I have a clear and distinct idea of Descartes ducking and dodging!

Here's my (somewhat cheeky) summary, with comments and evaluation at the end.

Elisabeth, May 6, 1643:

I'm so ignorant and you're so learned! Here's what I don't understand about your view: How can an immaterial soul, simply by thinking, possibly cause a bodily action?

Specifically,

it seems that how a thing moves depends solely on (i) how much it is pushed, (ii) the manner in which it is pushed, or (iii) the surface-texture and shape of the thing that pushes it. The first two of those require contact between the two things, and the third requires that the causally active thing be extended [i.e., occupy a region of space]. Your notion of the soul entirely excludes extension, and it appears to me that an immaterial thing can't possibly touch anything else.

Also, if, as you say, thinking is the essential property of human souls, what about unborn children and people who have fainted, who presumably have souls without thinking?

René, May 21, 1643:

Admittedly in my writings I talk much more about the fact that the soul thinks than about the question of how it is united with the body. This idea of the union of the soul and the body is basic and can be understood only through itself. It's so easy to get confused by using your imagination or trying to apply notions that aren't appropriate to the case!

For a comparison, however, think about how the weight of a rock moves it downwards. One might (mistakenly, I hope later to show) think of weight as a "real quality" about which we know nothing except that it has the power to move the body toward the centre of the earth. The soul's power to move the body is analogous.

Elisabeth, June 10, 1643:

Please forgive my stupidity! I wish I had the time to develop your level of expertise. But why should I be persuaded that an immaterial soul can move a material body by this analogy to weight? If we think in terms of the old idea of weight, why shouldn't we then conclude by your reasoning that things move downward due to the power of immaterial causes? I can't conceive of "what is immaterial" except negatively as "what is not material" and as what can't enter into causal relations with matter. I'd rather concede that the soul is material than that an immaterial thing could move a body.

René, May 28, 1643:

This matter of the soul's union with the body is a very dark affair when it comes from the intellect (whether alone or aided by the imagination). People who just use their senses, in the ordinary course of life, have no doubt that the soul moves the body. We shouldn't spend too much time in intellectual thinking. In fact,

I never spend more than a few hours a day in the thoughts involving the imagination, or more than a few hours a year on thoughts that involve the intellect alone. I give all the rest of my time to the relaxation of the senses and the repose of the mind.

The human mind can't clearly conceive the soul's distinctness from the body and its union with the body simultaneously. The comparison with weight was imperfect, but without philosophizing everyone knows that they have body and thought and that thought can move the body.

But since you remark that it is easier to attribute matter and extension to the soul than to credit it with the capacity to move and be moved by the body without having matter, please feel free to attribute this matter and extension to the soul -- because that's what it is to conceive it as united to the body.

Still, once you do this, you'll find that matter is not thought because the matter has a definite location, excluding other matter. But again, thinking too much about metaphysics is harmful.

Elisabeth, July 1, 1643:

I hope my letters aren't troubling you.

I find from your letter that the senses show me that the soul moves the body, but as for how it does so, the senses tell me nothing about that, any more than the intellect and imagination do. This leads me to think that the soul has properties that we don't know -- which might overturn your doctrine... that the soul is not extended.

As you have emphasized in your writings, all our errors come from our forming judgments about things we don't perceive well enough. Since we can't perceive how the soul moves the body, I am left with my initial doubt, that is, my thinking that perhaps after all the soul is extended.

There is no record of a reply by Descartes.

---------------------------------------

Zing! Elisabeth shows up so much better than Descartes in this exchange. She immediately homes in on the historically most important (and continuing) objection to Cartesian substance dualism: the question of how, if at all, an immaterial soul and a material object could causally interact. She efficiently and elegantly formulates a version of the principle of "the causal closure of the physical", according to which material events can only be caused by other material events, connecting that idea both with Descartes' denial that the soul is extended in space and with the view, widely accepted by early modern philosophers before Newton, that physical causation requires direct physical contact (no "action at a distance"). Jaegwon Kim notes (2011, p. 49) that hers might be the first causal argument for a materialist view of the mind. To top it off, she poses an excellent objection (from fetuses and fainting spells) to the idea that thinking is essential to having a soul.

Descartes' reply by analogy to weight is weak. As Elisabeth notes, it doesn't really answer the question of how the process is supposed to work for souls. Descartes' own theory of weight (articulated the subsequent year in Principles of Philosophy, dedicated to Elisabeth) involves action by contact (light particles spinning off the rotating Earth shoot up, displacing heavier particles down: IV.20-24). At best, Descartes is saying that the false, old idea of weight didn't involve contact, so why not think souls can also have influence without contact? Elisabeth's reply implicitly suggests a dilemma: If downward motion is by contact, then weight is not an example of how causation without contact is possible. If downward motion is not by contact, then shouldn't we think (absurdly?) that things move down due to the action of immaterial souls? She also notes that "immaterial" just seems to be a negative idea, not something we can form a clear, positive conception of.

Elisabeth's response forces Descartes concede that we can't in fact think clearly and distinctly about these matters. This is a major concession, given the centrality of the standard of "clear and distinct" ideas to Descartes' philosophy. He comes off almost as a mysterian! He also seems to partly retract what is perhaps the most central idea in his dualist metaphysics -- that the soul does not have extension. Elisabeth should feel free to attribute matter and extension to the soul, after all! Indeed, in saying that attributing matter and extension is "what it is to conceive [the soul] as united to the body", Descartes seriously muddies the interpretation of his positive view about the nature of souls.

It's also worth noting that Descartes entirely ignores Elisabeth's excellent fetus and fainting question.

I had previously been familiar with Descartes' famous quote that he spends no more than a few hours a year on thoughts involving the intellect alone; but reading the full exchange provides interesting context. His aim in saying that is to convince Elisabeth not to put too much energy into objecting to his account of how the soul works.

Understandably, Elisabeth is dissatisfied. She even gestures (though not in so many words) toward Descartes' methodological self-contradiction: Descartes famously says that philosophizing requires that we have clear ideas and that our errors all arise from failure to do so -- yet here he is, saying that there's an issue at the core of his metaphysics about which it's not possible to think clearly! Shouldn't he admit, then, that on this very point he's liable to be mistaken?

If Descartes attempted a further reply, the reply is lost. Their later correspondence treats other issues.

The whole correspondence is just 15 pages, so I'd encourage you to read it yourself. This summary necessarily omits interesting detail and nuance. In this exchange, Elisabeth is by far the better philosopher.

[image source]

Pierre Menard, Author of My ChatGPT Plagiarized Essay

If I use autocomplete to help me write my email, the email is -- we ordinarily think -- still written by me.  If I ask ChatGPT to generate an essay on the role of fate in Macbeth, then the essay was not -- we ordinarily think -- written by me.  What's the difference?

David Chalmers posed this question a couple of days ago at a conference on large language models (LLMs) here at UC Riverside.

[Chalmers presented remotely, so Anna Strasser constructed this avatar of him. The t-shirt reads: "don't hate the player, hate the game"]

Chalmers entertained the possibility that the crucial difference is that there's understanding in the email case but a deficit of understanding in the Macbeth case.  But I'm inclined to think this doesn't quite work.  The student could study the ChatGPT output, compare it with Macbeth, and achieve full understanding of the ChatGPT output.  It would still be ChatGPT's essay, not the student's.  Or, as one audience member suggested (Dan Lloyd?), you could memorize and recite a love poem, meaning every word, but you still wouldn't be author of the poem.

I have a different idea that turns on segmentation and counterfactuals.

Let's assume that every speech or text output can be segmented into small portions of meaning, which are serially produced, one after the other.  (This is oversimple in several ways, I admit.)  In GPT, these are individual words (actually "tokens", which are either full words or word fragments).  ChatGPT produces one word, then the next, then the next, then the next.  After the whole output is created, the student makes an assessment: Is this a good essay on this topic, which I should pass off as my own?

In contrast, if you write an email message using autocomplete, each word precipitates a separate decision.  Is this the word I want, or not?  If you don't want the word, you reject it and write or choose another.  Even if it turns out that you always choose the default autocomplete word, so that the entire email is autocomplete generated, it's not unreasonable, I think, to regard the email as something you wrote, as long as you separately endorsed every word as it arose.

I grant that intuitions might be unclear about the email case.  To clarify, consider two versions:

Lazy Emailer.  You let autocomplete suggest word 1.  Without giving it much thought, you approve.  Same for word 2, word 3, word 4.  If autocomplete hadn't been turned on, you would have chosen different words.  The words don't precisely reflect your voice or ideas, they just pass some minimal threshold of not being terrible.

Amazing Autocomplete.  As you go to type word 1, autocomplete finishes exactly the word you intend.  You were already thinking of word 2, and autocomplete suggests that as the next word, so you approve word 2, already anticipating word 3.  As soon as you approve word 2, autocomplete gives you exactly the word 3 you were thinking of!  And so on.  In the end, although the whole email is written by autocomplete, it is exactly the email you would have written had autocomplete not been turned on.

I'm inclined to think that we should allow that in the Amazing Autocomplete case, you are author or author-enough of the email.  They are your words, your responsibility, and you deserve the credit or discredit for them.  Lazy Emailer is a fuzzier case.  It depends on how lazy you are, how closely the words you approve match your thinking.

Maybe the crucial difference is that in Amazing Autocomplete, the email is exactly the same as what you would have written on your own?  No, I don't think that can quite be the standard.  If I'm writing an email and autocomplete suggests a great word I wouldn't otherwise have thought of, and I choose that word as expressing my thought even better than I would have expressed it without the assistance, I still count as having written the email.  This is so, even if, after that word, the email proceeds very differently than it otherwise would have.  (Maybe the word suggests a metaphor, and then I continue to use the metaphor in the remainder of the message.)

With these examples in mind, I propose the following criterion of authorship in the age of autocomplete: You are author to the extent that for each minimal token of meaning the following conditional statement is true: That token appears in the text because it captures your thought.  If you had been having different thoughts, different tokens would have appeared in the text.  The ChatGPT essay doesn't meet this standard: There is only blanket approval or disapproval at the end, not token-by-token approval.  Amazing Autocomplete does meet the standard.  Lazy Emailer is a hazy case, because the words are only roughly related to the emailer's thoughts.

Fans of Borges will know the story Pierre Menard, Author of the Quixote.  Menard, imagined by Borges to be a 20th century author, makes it his goal to authentically write Don Quixote.  Menard aims to match Cervantes' version word for word -- but not by copying Cervantes.  Instead Menard wants to genuinely write the work as his own.  Of course, for Menard, the work will have a very different meaning.  Menard, unlike Cervantes, will be writing about the distant past, Menard will be full of ironies that Cervantes could not have appreciated, and so on.  Menard is aiming at authorship by my proposed standard: He aims not to copy Cervantes but rather to put himself in a state of mind such that each word he writes he endorses as reflecting exactly what he, as a twentieth century author, wants to write in his fresh, ironic novel about the distant past.

On this view, could you write your essay about Macbeth in the GPT-3 playground, approving one individual word at a time?  Yes, but only in the magnificently unlikely way that Menard could write the Quixote.  You'd have to be sufficiently knowledgeable about Macbeth, and the GPT-3 output would have to be sufficiently in line with your pre-existing knowledge, that for each word, one at a time, you think, "yes, wow, that word effectively captures the thought I'm trying to express!"